Stroke is the leading cause of adult disability. Despite conventional therapy, a majority of stroke survivors have persistent hand dysfunction. This proposed research will further our understanding of mechanisms of recovery of hand function and provide a scientific basis for post-stroke rehabilitation protocols in clinical practie. Motor recovery occurs through motor learning, which depends on at least three processes: adaptation, repetition and reinforcement. Adaptation is the ability to predict forces and movements according to the expected consequences of an action, and is fundamental to motor skill learning. When adapted movements are reinforced through repetition, learning is enhanced. We have shown that adaptation of fingertip forces and movements is disrupted after stroke. However, we found that context-specific sensory information from the unaffected hand can restore adaptation in the affected hand. These results suggest that the 'good' hand can teach the 'bad' hand fundamental aspects of grasp control. We also found that increased activity in anti-gravity postural muscles is associated with increased efficiency of finger movements. Based on these results we hypothesize that an alternate hand practice strategy, enhanced with postural muscle activation, will improve the rate of motor re-learning and enhance recovery of hand function post stroke. This strategy will tap into redundant connectivity between the two sides of the brain to facilitate sensorimotor integration during skill learning. In Aim 1, we will refine the alternate hand practice strategy to restore adaptation in the affected hand. The experiments will lead to a better understanding of how kinesthetic, tactile and visual sensory modalities interact during adaptation, given existing sensorimotor impairments after stroke. In Aim 2, we will delineate postural strategies that improve grasp efficiency and reduce abnormal directional biases to facilitate repetition of more efficient movements during practice. In Aim 3, we will examine the rate of within-session and between- session motor learning and improvement in hand function with the 'enhanced alternate hand training paradigm', and compare this to training with the affected hand alone, using a novel task panel and structured training protocols. We will test our Aims using quantitative psychophysical measurements of fingertip forces, finger kinematics, 3-D arm and trunk kinematics, and electromyographic recordings of bilateral upper limb muscle activity during functional reach-to-grasp and lift tasks. We expect to identify impairments that directly affect function, provide objective information about how a task is performed, and inform treatment strategies and dosing of therapy for enhanced re-learning. This project will advance our understanding of how redundant circuitry can be harnessed for integration of sensory input with motor output for re-learning, and provide a scientific basis for the selection of content and structure of practice. The knowledge obtained will inform best- practice rehabilitation protocols for recovery of hand function, which will impac the quality of life of individuals after stroke and other neurological conditions.